WO2020048975A1 - Continuous preparation of an optically active carbonyl compound by asymmetric hydrogenation - Google Patents

Abstract

A process for continuous preparation of an optically active carbonyl compound by asymmetric hydrogenation of a prochiral α,β-unsaturated chiral compound with hydrogen in the presence of a homogeneous rhodium catalyst having at least one chiral ligand, wherein a liquid reaction mixture containing the prochiral α,β-unsaturated carbonyl compound is subjected to a biphasic gas/liquid hydrogenation in a first backmixed reactor, and the liquid reaction mixture is then further hydrogenated in a second reactor, wherein a concentration of the prochiral α,β-unsaturated carbonyl compound in the first reactor of 3% to 20% by weight is established. The process permits a high overall conversion to the prochiral unsaturated α,β-carbonyl compound.

Description

 Continuous production of an optically active carbonyl compound by asymmetric hydrogenation

The present invention relates to a process for the preparation of optically active carbonyl compounds by asymmetric hydrogenation of a prochiral a, b-unsaturated carbonyl compound with hydrogen in the presence of at least one homogeneous rhodium catalyst which has at least one chiral ligand. In particular, the present invention relates to a method for producing optically active citronellal.

Many optically active aldehydes and ketones are valuable intermediates for the synthesis of more highly refined chiral valuable substances and active ingredients and are in turn often sought-after fragrance and aroma substances.

Menthol is one of the most important aroma chemicals, the bulk of which is still isolated from natural sources. A totally synthetic approach for the naturally occurring enantiomer L-menthol uses the asymmetric (enantioselective) hydrogenation of geraniol or nerol to optically active citronellal. The optically active citronellal can then be cyclized with acid to L-isopulegol and hydrogenated to L-menthol. There is a constant need to optimize the economics of the process and especially the introductory asymmetric hydrogenation step.

DE 198 54 637 A1 describes a reactor for the continuous implementation of gas-liquid, liquid-liquid and gas-liquid-solid reactions, with a downwardly directed jet nozzle and a concentrically arranged guide tube.

EP-A 0 000 315 relates to a process for the production of optically active citronellal by hydrogenation of geranial or neral in the presence of a catalyst complex of rhodium and a chiral phosphine dissolved in the reaction system.

DE 10 2004 049 631 describes a process for the preparation of optically active carbonyl compounds by asymmetric hydrogenation of a, b-unsaturated carbonyl compounds in the presence of optically active transition metal catalysts which are soluble in the reaction system and which have at least one carbon monoxide ligand. WO 2009/153123 describes a continuous process for the hydrogenation of organic compounds in a multiphase system in the presence of a homogeneous or heterogeneous catalyst, the process being carried out in two stages, the first stage in a loop reactor with an external heat exchanger and the second stage in one Bubble column reactor with limited backmixing is carried out.

The present invention relates to a process for the continuous production of an optically active carbonyl compound by asymmetric hydrogenation of a prochiral a, b-unsaturated carbonyl compound with hydrogen in the presence of a homogeneous rhodium catalyst which has at least one chiral ligand, the prochiral a, b- unsaturated carbonyl compound-containing liquid reaction mixture in a first, back-mixed reactor is subjected to a gas / liquid two-phase hydrogenation, and the liquid reaction mixture is then further hydrogenated in a second reactor, a concentration of the prochiral a, b-unsaturated carbonyl compound of 3 being in the first reactor sets up to 20 wt .-%.

It was found that the reaction rate of the hydrogenation of the prochiral a, b-unsaturated carbonyl compound with hydrogen in the presence of the homogeneous rhodium catalyst initially increases steeply with the concentration of the a, b-unsaturated carbonyl compound at low concentrations of the a, b-unsaturated carbonyl compound, but then sharply decreases at higher concentrations of the a, b-unsaturated carbonyl compound. Presumably the olefinic double bond of the a, b-unsaturated carbonyl compound can interact with the rhodium atom of the catalyst at high concentrations of the a, b-unsaturated carbonyl compound and reversibly inhibit the catalyst. Such behavior is also referred to as educt inhibition.

In the process according to the invention, a liquid reaction mixture is reacted in a first, back-mixed reactor. Back-mixing with the hydrogenation product leads to a dilution of the a, b-unsaturated carbonyl compound and a minimization of the educt inhibition.

The reaction rate can be expressed as the amount of a, b-unsaturated carbonyl compound converted per time, normalized to the amount of substance Rhodium atoms. When applied (under the pressure and temperature conditions in the first reactor), it shows a maximum V _max against the concentration of the prochiral a, b-unsaturated carbonyl compound. A concentration of the prochiral a, b-unsaturated carbonyl compound at which the reaction rate is at least 0.8 times V _max is preferably set in the first reactor.

A concentration of the prochiral a, b-unsaturated carbonyl compound in the liquid phase of 3 to 20% by weight, preferably 5 to 20% by weight, in particular 5 to 15% by weight, such as 7 to 13% by weight, is set in the first reactor .-%, 8 to 12 wt .-% or 9 to 1 1 wt .-%, for example 10 wt .-%. Since, despite the backmixing in the first reactor, the concentration of the prochiral a, b-unsaturated carbonyl compound in the liquid phase is not completely homogeneous, the concentration of the prochiral a, b-unsaturated carbonyl compound is the concentration in the liquid phase at the outlet of the first reactor.

In order to increase sales, the liquid reaction mixture is then reacted further in a second reactor. High concentrations of unreacted prochiral a, b-unsaturated carbonyl compound (hereinafter also referred to as "educt") in the final reaction mixture also lead to high residual concentrations of prochiral a, b-unsaturated carbonyl compound in the catalyst residue, from which the optically active carbonyl compound z. B. is separated by distillation. The catalyst residue is usually returned to the hydrogenation reaction after a preforming described below. High residual concentrations of prochiral a, b-unsaturated carbonyl compound impair preformation. It is therefore desirable to largely complete the conversion in the second reactor and to minimize the concentration of unreacted prochiral a, b-unsaturated carbonyl compound.

The liquid reaction mixture is preferably set in the second reactor to a concentration of the prochiral a, b-unsaturated carbonyl compound of less than 5% by weight, in particular less than 3% by weight, such as less than 2% by weight or less than 1 wt%.

The ratio of the reaction volume of the first reactor to the reaction volume of the second reactor is preferably 1: 1 to 1: 5, in particular 1: 2 to 1: 4, for example approximately 1: 3.5. The residence time behavior in a reactor can be represented by the cascade model (see, for example, Baerns, Hofmann, Renken, "Chemical Reaction Technology"), which describes a real reactor by a theoretical number N (so-called number of vessels) of ideal stirred vessels connected in series. It applies that the boilers have the same volume and the reaction volume does not change. A backmixing between the individual boilers is impossible. The boiler number N can be between 1 and infinity, but may only take whole values. The extreme conditions are N = 1 (complete mixing, ideal stirred tank) or N = (no axial backmixing, ideal flow tube).

In the process according to the invention, the first reactor preferably has a number of vessels in the range from 1 to 3, in particular 1 or 2.

The first reactor is preferably designed as a loop reactor. Examples of loop reactors are tubular reactors with internal and external loops. Such reactors are described in more detail, for example, in Ullmann's Encyclopedia (Ullmann's Encyclopedia of Industrial Chemistry, Verlag Chemie, Electronic Release 2008, 7th Edition, chapter "Stirred Tank and Loop Reactors" and chapter "Bubble Columns"). A loop reactor usually consists of a vertical, preferably cylindrical, tubular reactor.

The ratio of length to diameter of the loop reactor is usually 2: 1 to 100: 1, preferably 5: 1 to 100: 1, particularly preferably 5: 1 to 50: 1, particularly preferably 5: 1 to 30: 1.

The educt is fed in at any point in the tubular reactor, preferably through a jet nozzle, in particular a jet nozzle arranged above the liquid level. The term jet nozzle usually refers to a tube that tapers in the direction of flow. The jet nozzle mixes the first reactor.

The volume-specific power input into the first reactor is preferably 0.5 to 5 kW / m ³ , for example 1.0 to 4 kW / m ³ or 1.5 to 3 kW / m ³ . The volume-specific power input can be determined as the product of the differential pressure across the jet nozzle and the volume flow through the nozzle.

The jet nozzle can be designed as a single-substance or two-substance nozzle. In the single-component nozzle, only the liquid reaction mixture is injected and the hydrogen at any other point, but preferably in the lower region of the reactor. The simple construction of such a single-substance nozzle is advantageous in this embodiment. The injection of hydrogen in the lower region of the reactor advantageously supports the circulation of the reaction mixture in the reactor. In the two-component nozzle, the hydrogen is fed in and dispersed together with the liquid reaction mixture.

In a preferred embodiment, the reaction mixture is metered down via a single-component jet nozzle located at the top of the reactor. Hydrogen collects in the gas space above the liquid level in the reactor. The jet from the nozzle falls through the gas space and, when entering the liquid phase, strikes the hydrogen gas, which is broken down into bubbles and dispersed in the liquid phase.

The loop reactor is generally designed as a tubular reactor with an external circuit (external loops). In the case of a loop reactor with an external circuit, there is generally an outlet at any point in the reactor, preferably in the lower region of the reactor, via which the reaction mixture is fed back to the jet nozzle in an external circuit by means of a delivery device. The delivery device is preferably a pump, which is why the external circuit is usually referred to as a pumping circuit.

Examples of pumps are centrifugal pumps or rotary piston pumps, such as rotary lobe pumps, rotary vane pumps, rotary lobe pumps or gear pumps. Centrifugal pumps are particularly preferably used as the delivery device.

The first reactor is preferably designed as a loop reactor with an external circuit, a heat exchanger being located in the external circuit. Such a reactor is referred to in the context of this invention as a loop reactor with an external heat exchanger.

The heat exchanger is, for example, a tube bundle heat exchanger, double tube heat exchanger, plate heat exchanger or spiral heat exchanger. At design pressures for the reactor below 100 bar, a shell-and-tube heat exchanger is preferably used; at higher pressures, one or more double-tube heat exchangers connected in series are preferably used.

The loop reactor with an external heat exchanger is usually operated in such a way that part of the reaction mixture is conveyed out of the reactor through the external pumping circuit in which the external heat exchanger is located. wherein the reaction mixture conveyed by the heat exchanger is cooled. External pumping generally mixes and circulates the reaction mixture vigorously in the first reaction stage, so that the residence time in the first stage usually corresponds to that of a completely backmixed stirred tank (CSTR).

The reaction mixture is finally fed back to the reactor by means of the jet nozzle. Fresh starting materials and catalyst solution are usually metered into the pumping circuit and fed to the reactor as a reaction mixture together with the stream already in the pumping circuit.

In a preferred embodiment, the loop reactor is designed such that a so-called internal loop flow is formed in addition to the external circuit. In a loop reactor with an internal loop flow, a concentric, preferably cylindrical guide tube is generally arranged in the interior of the tube reactor and extends essentially over the entire length of the tube reactor with the exception of the reactor ends.

The guide tube is usually designed as a simple tube. The ratio of length to diameter of the guide tube is generally 5: 1 to 100: 1, preferably 5: 1 to 50: 1.

The diameter of the guide tube is less than the diameter of the tube reactor. The ratio of the diameter of the guide tube to the diameter of the tube reactor is generally 0.3: 1 to 0.9: 1, preferably 0.5: 1 to 0.7: 1. The space between the guide tube and the inner wall of the reactor is generally referred to as an annular space.

The jet nozzle is usually arranged such that the gas / liquid jet generated by the jet nozzle is directed into the guide tube. The jet nozzle is preferably arranged above the upper end of the guide tube. The nozzle tip is located above the liquid level and does not dip into the liquid phase. The gas / liquid jet generated by the jet nozzle causes a downward flow in the guide tube (outflow column), which is deflected after leaving the guide tube, so that the liquid in the annular space between the guide tube and the inner wall of the reactor flows again in the direction of the jet nozzle (upstream column). This usually creates an internal loop flow. The ratio of the volume flows from internal loop flow to externally pumped reaction mixture is preferably 2 to 30: 1, particularly preferably 5 to 20: 1. At least part of the reaction mixture is fed to the second reactor from the first reactor. Backmixing in the second reactor is preferably limited, so that the residence time distribution in this second reactor approximates that of a tubular reactor. Due to this defined liquid retention time, the conversion of the starting material can take place almost completely. The second reactor preferably has a number of vessels of more than 4, in particular more than 5 or more than 6.

The second reactor usually consists of a vertical, preferably cylindrical tubular reactor. The ratio of length to diameter of the reactor is usually 2: 1 to 100: 1, preferably 5: 1 to 50: 1, particularly preferably 7: 1 to 25: 1.

In a section of the second reactor which is located at the outlet from the second reactor, the hydrogenation preferably takes place in liquid-single phase, ie. H. there is no dispersed gas phase in the section located at the outlet from the second reactor and the hydrogenation is carried out exclusively with the hydrogen dissolved in the liquid phase. Since the hydrogenation conversions towards the exit from the second reactor are low, the concentration of dissolved hydrogen is sufficient. Due to the absence of a discrete gas phase for the exit from the second reactor, the liquid holdup of the second reactor can be increased and the residence time of the liquid phase in the second reactor can be increased. Since the hydrogenation takes place in the liquid phase, the reaction space is optimally used in this way. The liquid-phase section of the second reactor preferably accounts for 30 to 50% of the total volume of the second reactor.

Backmixing in the second reactor is preferably limited by internals. By installing such devices, the circulation and thus the backmixing of gas and liquid are generally restricted.

The backmixing in the second reactor can be limited by installing various internals. In a preferred embodiment, the backmixing is limited by the installation of several, fixedly arranged trays in the tubular reactor. Thus, individual segments ("compartments") with defined reaction volumes are created between the individual trays. Each of the individual segments usually acts like a single, back-mixed stirred tank reactor. With an increasing number of individual segments in succession, the residence time distribution of such a cascade generally approaches the residence time of a tubular reactor. The number of individual segments formed in this way is preferably 2 to 20, particularly preferably 2 to 10, particularly preferably 3 to 6. The floors are preferably designed as liquid-permeable floors. The bottoms are particularly preferably designed as perforated plates.

In order to ensure sufficient hydrogen saturation, dispersion is preferably carried out in a section of the second reactor which is located at the inlet to the second reactor, e.g. B. in the first individual segment of the second reactor, hydrogen gas in the liquid reaction mixture.

For this purpose, the section of the second reactor which is located at the inlet to the second reactor is preferably designed as a loop reactor with an external circuit. The dispersion of hydrogen in the section is preferably carried out by a jet nozzle. The jet nozzle can be designed as a single-substance or two-substance nozzle. There is a fume hood at any point in the section via which the reaction mixture is fed back to the jet nozzle in an external circuit by means of a conveying element. The hydrogen can be fed in together with the liquid reaction mixture or metered at any point into the section located for entry into the second reactor.

The loop reactor with an external circuit preferably comprises a heat exchanger and is operated in such a way that the reaction mixture led from the reactor through the external pumping circuit is cooled by the heat exchanger.

In a preferred embodiment, the reaction mixture is metered down via a single-substance jet nozzle located at the top of the second reactor. The jet from the nozzle falls through the gas space and, when entering the liquid phase, strikes the gas, which is broken up into bubbles and dispersed in the liquid phase.

The number of individual segments of the second reactor is preferably 2 to 10, particularly preferably 2 to 10, particularly preferably 3 to 6. The volume of the first individual segment in which hydrogen gas is dispersed in the liquid reaction mixture is preferably 30 to 70%, based on the total volume of the second reactor. The second to last individual segments of the second reactor are preferably operated in liquid single phase; the second to last individual segment of the second reactor then preferably accounts for 30 to 70% of the total volume of the second reactor. The first reactor and the second reactor are preferably arranged as two spatially separate apparatuses which are connected to one another via pipelines. In other embodiments, the two reactors can be arranged in one apparatus (hydrogenation reactor). In this preferred embodiment, the hydrogenation reactor is designed as a long, cylindrical tube (high-cylindrical design). A prochiral a, b-unsaturated carbonyl compound can form a chiral center by addition reaction to the olefinic double bond. For this purpose, the double bond carries four different substituents. The prochiral a, b-unsaturated carbonyl compound is preferably selected from compounds of the general formula (I)

 (i)

 wherein

R ¹ , R ² are different from one another and each represent an unbranched, branched or cyclic hydrocarbon radical having 1 to 25 carbon atoms, which is saturated or has one or more, non-conjugated ethylenic double bonds and which is unsubstituted or one or more identical or different Carries substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms,

R ^{3 represents} hydrogen or an unbranched, branched or cyclic hydrocarbon radical having 1 to 25 carbon atoms, which is saturated or has one or more non-conjugated ethylenic double bonds and which is unsubstituted or carries one or more identical or different substituents which are selected are under OR ⁴ , NR ^5a R ^5b , halogen, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms, or

R ³ together with one of the radicals R ¹ or R ² also has a 3 to 25-membered group

Can mean alkylene group in which 1, 2, 3 or 4 non-adjacent CH2 groups can be replaced by O or NR ^5c , the Alkylene group is saturated or has one or more, non-conjugated ethylenic double bonds, and wherein the alkylene group is unsubstituted or bears one or more identical or different substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, Ci-C ₄ alkyl , C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms, where two substituents can also together represent a 2 to 10-membered alkylene group, the 2 to 10-membered alkylene group being saturated or one or more, non-conjugated ethylenic Has double bonds, and wherein the 2- to 10-membered alkylene group is unsubstituted or carries one or more identical or different substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms; in which

R ^{4 is} hydrogen, Ci-C ₆ alkyl, C ₆ -Cio-aryl, C6-Ci4-aryl-Ci-Cio-alkyl, or

 Ci-Cio-alkyl-C6-Ci4-aryl-;

R ⁵³ , R ^5b each independently of one another are hydrogen, C ₁ -C ₆ -alkyl, C ₆ -cio-aryl,

 C6-Ci4-aryl-Ci-Cio-alkyl or Ci-Cio-alkyl-C6-Ci4-aryl mean or

R ^5a and R ^5b together can also mean an alkylene chain with 2 to 5 carbon atoms, which can be interrupted by N or O; and

R ^5c for hydrogen, Ci-C ₆ alkyl, C ₆ -Cio-aryl, C6-Ci4-aryl-Ci-Cio-alkyl or

 Ci-Cio-alkyl-C6-Ci4-ary- stands.

In preferred embodiments, the prochiral a, b-unsaturated carbonyl compound is selected from compounds of the general formulas (la) and (Ib)

worin

wherein

worin ^* das Asymmetriezentrum bezeichnet; durch asymmetrische Hydrierung von Geranial der Formel (la-1 ) oder von Neral der Formel (lb-1 ) R ¹ , R ² each represent an unbranched or branched hydrocarbon radical having 2 to 25 carbon atoms, which is saturated or has 1, 2, 3, 4 or 5 non-conjugated ethylenic double bonds. A particularly preferred embodiment relates to a process for the production of optically active citronellal of the formula (III)

where ^* denotes the center of asymmetry; by asymmetric hydrogenation of geranial of the formula (la-1) or of neral of the formula (lb-1)

oder einer Neral und Geranial enthaltenden Mischung. Eine Neral und Geranial enthaltende Mischung ist unter der Bezeichnung Citral bekannt.

or a mixture containing neral and geranial. A mixture containing neral and geranial is known under the name Citral.

The optically active citronellal of the formula (III) thus obtainable can be subjected to a cyclization to optically active isopulegol and the optically active isopulegol can be hydrogenated to optically active menthol.

The method according to the invention makes it optically active

To provide carbonyl compounds, especially optically active aldehydes in high yields and enantiomeric excesses. The desired asymmetrically hydrogenated compounds are usually obtained in an enantiomeric excess of at least 80% ee, often with an enantiomeric excess of about 85 to about 99% ee. It should be noted that the maximum achievable enantiomeric excess can depend on the purity of the substrate used, particularly with regard to the isomeric purity of the double bond to be hydrogenated. Accordingly, suitable starting substances are in particular those which have an isomer ratio of at least about 90:10, preferably at least about 95: 5, with respect to the E / Z double bond isomers.

The production process according to the invention is carried out in the presence of an optically active rhodium catalyst which is soluble in the reaction mixture and which has at least one optically active ligand. Such catalysts can be obtained, for example, by reacting a suitable rhodium compound which is soluble in the reaction mixture with an optically active ligand which has at least one phosphorus and / or arsenic atom.

Examples of rhodium compounds which can be used according to the invention are: RhC, Rh (OAc) ₃ , [Rh (cod) CI] ₂ , Rh (CO) ₂ acac, [Rh (cod) OH] ₂ , [Rh (cod) OMe] ₂ , Rh ₄ (CO) i ₂ , Rh ₆ (CO) i ₆ , where "acac" stands for an acetylacetonate and "cod" for a cyclooctadiene ligand.

The catalyst concentration in the reaction mixture is preferably 0.001 to 1 mol%, in particular 0.002 to 0.5 mol%, particularly preferably 0.005 to 0.2 mol%, based on the amount of prochiral a, b-unsaturated carbonyl compound in the reaction mixture as rhodium atoms contained in the catalyst.

The rhodium compounds mentioned are brought into contact with a further compound which is optically active, preferably essentially enantiomerically pure (i.e. has an enantiomeric excess of at least about 99%) and has at least one phosphorus and / or arsenic atom, preferably at least one phosphorus atom. This compound, which can be referred to as a chiral ligand, forms the rhodium catalyst to be used according to the invention in the reaction mixture or in the preforming mixture with the rhodium compound used.

Those chiral ligands which have two phosphorus atoms and form chelate complexes with rhodium are particularly preferred.

Suitable chiral ligands in the context of the present invention are compounds such as those described, for example, in: I. Ojima (ed.), Catalytic Asymmetry Synthesis, Wiley-VCh, 2nd edition, 2000 or in EN Jacobsen, A. Pfaltz, H. Yamamoto (ed.), Comprehensive Asymmetrie Catalysis, 2000, Springer or in W. Tang, X. Zhang, Chem. Rev. 2003, 103, 3029-3069. Preferred ligands are chiral bidentate bisphosphine ligands, in particular those of the general formulas (IV) to (VI)

wherein

R ⁵ , R ⁶ each independently represent an unbranched, branched or cyclic hydrocarbon radical having 1 to 20 carbon atoms, which is saturated or can have one or more, generally 1 to about 4, non-conjugated, ethylenic double bonds and which is unsubstituted is or carries one or more, usually 1 to 4, identical or different substituents which are selected from OR ¹³ , NR ¹⁴ R ¹⁵ , halogen, C ₆ -Cio-aryl and Cs-Cg-hetaryl, or

R ⁵ and R ⁶ together can mean a 2 to 10-membered alkylene group or a 3 to 10-membered cycloalkylene group, in which 1, 2, 3 or 4 non-adjacent CH groups can be replaced by O or NR ¹³ , the alkylene group and the cycloalkylene group is saturated or has one or two non-conjugated ethylenic double bonds, and the alkylene group and the cycloalkylene group are unsubstituted or carry one or more identical or different substituents which are selected from C 1 -C ₄ -alkyl;

R ⁷ , R ⁸ each independently represent hydrogen or straight-chain or branched Ci-C ₄ alkyl and

R ⁹ , R ¹⁰ , R ¹¹ , R ^{12 are} identical or different and represent C ₆ -Cio-aryl which is unsubstituted or carries one or more substituents which are selected from Ci-C ₆ -alkyl, C3-C6- Cycloalkyl, C ₆ -Cio aryl, Ci-C ₆ alkoxy and amino;

R ¹³ , R ¹⁴ , R ¹⁵ each independently represent hydrogen, Ci-C ₄ alkyl, C ₆ -Cio aryl, C7-Ci2-aralkyl or C7-Ci2-alkylaryl, where R ¹⁴ and R ¹⁵ also together Alkylene chain with 2 to 5 carbon atoms, which can be interrupted by N or O, can mean.

With regard to formulas (IV), (V) and (VI), the variables have the following meanings in particular:

R ⁵ , R ⁶ each independently represent C 1 -C ₄ -alkyl or

R ⁵ and R ⁶ together represent a Cs-Cs-alkanediyl radical, C3-C7-alkenediyl radical, Cs-C7-cycloalkanediyl radical or a C5-C7-cycloalkenediyl radical, the four aforementioned radicals being unsubstituted or carry one or more identical or different substituents which are selected from C 1 -C ₄ -alkyl;

R ⁷ , R ⁸ are each independently hydrogen or C 1 -C ₄ -alkyl;

R ⁹ , R ¹⁰ , R ¹¹ , R ¹² each represent phenyl.

Because of their easy availability, particularly preferred chiral, bidentate bisphosphine ligands are compounds of the formula obtainable under the name "Chiraphos":

The chiral ligands are advantageously used in an amount of about 0.9 to about 10 mol, preferably about 1 to about 4 mol, per mol of rhodium compound used. The optically active rhodium catalyst which is soluble in the reaction mixture is suitably generated in-situ before or during the hydrogenation by reacting an achiral rhodium compound and with a chiral, bidentate bisphosphine ligand. In this context, the term "in-situ" means that the catalyst is generated directly before or at the start of the hydrogenation. The catalyst is preferably generated before the hydrogenation. It has been found that the presence of monodentate ligands can increase the activity of the catalyst. In a preferred embodiment of the process according to the invention, compounds of the formula (II)

17th

 R

eingesetzt, worin Z in Formel (II) für eine Gruppe CHR¹⁸R¹⁹ steht und worin die Variablen R¹⁶, R¹⁷, R¹⁸, R¹⁹ unabhängig voneinander und insbesondere gemeinsam die folgenden Bedeutungen aufweisen:

used, in which Z in formula (II) represents a group CHR ¹⁸ R ¹⁹ and in which the variables R ¹⁶ , R ¹⁷ , R ¹⁸ , R ¹⁹ independently and in particular together have the following meanings:

R ¹⁶ , R ¹⁷ : are identical or different and stand for phenyl which is unsubstituted or carries 1, 2 or 3 substituents which are selected from methyl and methoxy, where R ¹⁶ and R ¹⁷ each stand in particular for unsubstituted phenyl;

R ¹⁸ represents C to C ₄ alkyl, in particular methyl;

R ¹⁹ stands for C to C ₄ alkyl which carries a group P (= 0) R ^19a R ^19b , and in particular a group CH ₂ -P (= 0) R ^19a R ^19b or CH (CH ₃ ) - P (= 0) R ^{19a means} R ^19b ; in which

R ^i9a , R ^19b : are identical or different and represent phenyl which is unsubstituted or carries 1, 2 or 3 substituents which are selected from methyl and

Methoxy, R ^19a and R ^19b each particularly preferably representing unsubstituted phenyl.

In this preferred embodiment of the process according to the invention, a compound of the formula (II) in which

R ¹⁶ , R ¹⁷ : stand for unsubstituted phenyl;

R1S represents methyl; R ^{19 represents} a group CH (CH ₃ ) -P (= 0) R ^19a R ^19b , where R ^19a and R ^19b each represent unsubstituted phenyl.

This is the compound (2- (diphenylphosphoryl) -1-methylpropyl) diphenylphosphine (chiraphosmonooxide), including the (R, R) enantiomer (= (/?, / X) chiraphosmonooxide) and (S, S) enantiomers (= (S, S) -

Chiraphosmonooxid) as well as mixtures of (/?, / X) -Chiraphosmonooxid and (S, S) - Chiraphosmonooxid.

If the radicals R ¹⁸ and R ¹⁹ in the general formula (II) have different meanings, the carbon atom which carries the radicals R ¹⁸ and R ^{19 can have} an (R) or (S) configuration. These compounds of the general formula (II) can exist as pure (R) or pure (S) stereoisomers or as mixtures thereof. As a rule, the pure (R) and (S) stereoisomers will be used in these cases, and any stereoisomer mixtures are also suitable for use in the present process.

A pure stereoisomer is understood here and below to mean chiral substances which have an enantiomeric excess (ee) of at least 80% ee, in particular at least 90% ee and especially at least 95% ee with respect to the desired stereoisomer.

In particular, Chiraphos is used as the chiral ligand and (2- (diphenylphosphoryl) -1-methylpropyl) diphenylphosphine (chiraphos monoxide) as the monodentate binding compound. For example, R-Chiraphos is used as the chiral ligand and (/?, / X) -chiraphosmonooxide and / or (S, S) -chiraphosmonooxide is used as the monodentate binding compound. Alternatively, S-Chiraphos is used as the chiral ligand and (/?, / X) -chiraphosmonooxide and / or (S, S) -chiraphosmonooxide as the monodentate binding compound.

According to the invention, the compound of the formula (II) is generally used in an amount of 0.01 to 1 mol, preferably 0.02 to 0.8 mol, particularly preferably 0.03 to 0.7 mol and in particular in an amount of 0 , 04 to 0.6 moles per mole of rhodium.

Further embodiments of the rhodium catalyst and the monodentate ligand are described in US 2018/0057437 A1, WO 2006/040096 A1 and WO 2008/132057 A1. The hydrogen used for the hydrogenation is generally used in a larger stoichiometric excess of from 1 to 10 times, preferably from 1.1 to 5 times, the stoichiometrically necessary amounts. It can be returned to the reaction as cycle gas. The hydrogen is generally used in a technically pure manner. The hydrogen can also be used in the form of a gas containing hydrogen, ie in admixtures with other inert gases such as nitrogen, helium, neon, argon or carbon dioxide. For example, reformer exhaust gases, refinery gases, etc. can be used as the gases containing hydrogen. However, pure hydrogen or essentially pure hydrogen is preferably used in the process.

The asymmetric hydrogenation according to the invention is advantageously used at a pressure of about 2 to about 200 bar, in particular from about 10 to about 100 bar, especially at about 60 to about 100 bar and a temperature of generally about 0 ° C. to about 100 ° C, preferably about 0 ° C to about 30 ° C, especially at about 10 ° C to about 30 ° C.

The choice of the solvent to be used to carry out the asymmetric hydrogenation according to the invention is of lesser importance. Suitable solvents or solvent media which are inert under the reaction conditions are, for example, ether, tetrahydrofuran, methyltetrahydrofuran, toluene, xylenes, chlorobenzene, octadecanol, biphenyl ether, texanol, Marlotherm, oxo oil 9N (hydroformylation products from isomeric octenes, BASF SE), citronellal and the like. The hydrogenation product or any high-boiling by-products that may occur during the reaction can also serve as the solution medium. The asymmetric hydrogenation is particularly advantageously carried out in the same solvent as any preforming which may have been carried out beforehand.

In a preferred embodiment of the process according to the invention, the catalyst is pretreated with a gas mixture containing carbon monoxide and hydrogen before the hydrogenation and / or the asymmetric hydrogenation is carried out in the presence of carbon monoxide additionally added to the reaction mixture.

This means that either the rhodium catalyst which is soluble in the reaction mixture, ie homogeneous, is pretreated either before the asymmetric hydrogenation with a gas mixture which contains carbon monoxide and hydrogen (ie carries out a so-called preforming) or the asymmetrical hydrogenation in the presence of the reaction mixture carries out additionally supplied carbon monoxide or carries out a preforming and then the asymmetrical Hydrogenation in the presence of carbon monoxide additionally supplied to the reaction mixture.

The rhodium catalysts which are soluble in the reaction mixture and are used in this preferred embodiment can therefore have at least one CO ligand, at least in a form which has passed through in the catalytic cycle or in a preform preceding the actual catalytic cycle, it being irrelevant whether these at least one CO Ligand-containing catalyst form represents the actual catalytically active catalyst form. In order to stabilize the catalyst forms which may have CO ligands, it may be advantageous to additionally add carbon monoxide to the reaction mixture during the hydrogenation.

In this preferred embodiment, the aforementioned pretreatment of the catalyst precursor is carried out with a gas mixture comprising 20 to 90% by volume of carbon monoxide, 10 to 80% by volume of hydrogen and 0 to 5% by volume of further gases, the volume percentages mentioned being 100 Vol .-% add, carried out at a pressure of 5 to 100 bar. In addition, excess carbon monoxide is separated off from the catalyst thus obtained before use in the asymmetric hydrogenation. The term excess carbon monoxide is understood to mean carbon monoxide which is contained in the reaction mixture obtained in gaseous or dissolved form and is not bound to the rhodium catalyst or its precursor. Accordingly, the excess carbon monoxide not bound to the catalyst is at least largely removed, i.e. to the extent that any residual amounts of dissolved carbon monoxide do not have a disturbing effect in the subsequent hydrogenation. This is usually ensured if about 90%, preferably about 95% or more, of the carbon monoxide used for the preforming is separated off. The excess carbon monoxide is preferably removed completely from the catalyst obtained by preforming.

The excess carbon monoxide can be separated off from the catalyst obtained or from the reaction mixture containing the catalyst in various ways. Preferably, the catalyst or the mixture obtained by preforming and containing the catalyst is depressurized to a pressure of up to about 5 bar (absolute), preferably, especially when the preforming is carried out in the pressure range from 5 to 10 bar, to a pressure from less than 5 bar (absolute), preferably to a pressure in the range from about 1 bar to about 5 bar, preferably 1 to less than 5 bar, particularly preferably to a pressure in the range from 1 to 3 bar, very particularly preferably to one Pressure in the range from about 1 to about 2 bar, particularly preferably to normal pressure, so that gaseous, unbound carbon monoxide escapes from the product of the preforming. The above-mentioned expansion of the preformed catalyst can be carried out, for example, using a high-pressure separator, as is known to the person skilled in the art. Such separators in which the liquid is in the continuous phase are described, for example, in: Perry's Chemical Engineers' Handbook, 1997, 7th edition, McGraw-Hill, pp. 14.95 and 14.96; the prevention of a possible droplet drought is described on pages 14.87 to 14.90. The preformed catalyst can be expanded in one or two stages until the desired pressure is reached in the range from 1 bar to about 5 bar, the temperature usually falling to 10 to 40 ° C. Alternatively, excess carbon monoxide can be separated off by stripping the catalyst or the mixture containing the catalyst with a gas, advantageously with a gas which is inert under the reaction conditions. The term stripping is understood by the person skilled in the art to mean the introduction of a gas into the catalyst or the reaction mixture containing the catalyst, for example in WRA Vauck, HA Müller, Basic Operations of Chemical Process Engineering, German Publisher for Basic Chemicals Leipzig, Stuttgart, 10th Edition, 1984, page 800 , described. Examples of suitable inert gases are: hydrogen, helium, neon, argon, xenon, nitrogen and / or CO 2, preferably hydrogen, nitrogen, argon.

The asymmetric hydrogenation is then preferably carried out using hydrogen which has a carbon monoxide content in the range from 50 to 3000 ppm, in particular in the range from 100 to 2000 ppm, especially in the range from 200 to 1000 ppm and very particularly in the range from 400 to 800 ppm.

If the rhodium catalyst is preformed, the selected rhodium compound and the selected chiral ligand are usually dissolved in a suitable solvent or solvent medium which is inert under the reaction conditions, such as, for example, ether, tetrahydrofuran, methyltetrahydrofuran, toluene, xylenes, chlorobenzene, octadecanol and biphenyl ether , Texanol, Marlotherm, Oxoöl 9N (hydroformylation products from isomeric octenes, BASF Aktiengesellschaft), citronellal and the like. The hydrogenation product or any high-boiling by-products that may occur during the reaction can also serve as the solution medium. The resulting solution, advantageously in a suitable pressure reactor or autoclave, at a pressure of usually about 5 to about 350 bar, preferably from about 20 to about 200 bar and particularly preferably from about 50 to about 100 bar, is injected with a gas mixture, the hydrogen and contains carbon monoxide. It is preferred to use a gas mixture for preforming, which is approximately 30 to 99% by volume of hydrogen,

 1 to 70 vol .-% carbon monoxide and

 Contains 0 to 5% by volume of other gases, the details in% by volume adding up to 100% by volume.

A particularly preferred gas mixture for preforming is so-called synthesis gas, which usually consists of about 35 to 55% by volume of carbon monoxide in addition to hydrogen and traces of other gases.

The preforming of the catalyst is usually carried out at temperatures from about 25 ° C. to about 100 ° C., preferably at about 40 ° C. to about 80 ° C. Preforming is usually complete after about 1 hour to about 24 hours, often after about 1 to about 12 hours. Following the optional preforming, the asymmetric hydrogenation of the selected substrate is carried out according to the invention. After the previous preforming, the selected substrate can generally be carried out successfully with or without the addition of additional carbon monoxide. If no preforming is carried out, the asymmetric hydrogenation according to the invention can be carried out either in the presence of carbon monoxide fed to the reaction system or without the addition of carbon monoxide. A preforming as described is advantageously carried out and additional carbon monoxide is added to the reaction mixture during the asymmetric hydrogenation.

If carbon monoxide is fed to the reaction system, the feed can be carried out in various ways: For example, the carbon monoxide can be mixed with the hydrogen used for the asymmetric hydrogenation or can be metered directly into the reaction solution in gaseous form. The carbon monoxide is preferably admixed with the hydrogen used for the asymmetric hydrogenation.

The reaction product can be removed from the reaction mixture by processes known per se to the person skilled in the art, such as, for example, by distillation and / or flashing evaporation, and the remaining catalyst, if appropriate after repeated preforming, can be used in the course of further reactions. In the context of the preferred embodiment, the addition of solvents is advantageously dispensed with and the reactions mentioned are carried out in the substrate or the product to be reacted and, if appropriate, in high-boiling by-products as the dissolving medium. Continuous reaction control is particularly preferred with reuse or recycling of the homogeneous catalyst stabilized according to the invention.

The invention is illustrated in more detail by the attached figures and the following example.

1 shows a plot of the reaction rate (relative to the maximum reaction rate) of the hydrogenation of citral in the presence of a chiraphos rhodium catalyst as a function of the citral concentration.

2 schematically shows a plant suitable for carrying out the method according to the invention.

3 schematically shows a plant suitable for carrying out the method according to the invention.

The data shown in FIG. 1 were obtained in a jet loop reactor in which different conversions were achieved due to different catalyst loads. The difference between the amount of citral entering and exiting was related to the mass of rhodium present in the reactor and plotted against the citral concentration in the discharge from the reactor. This reaction rate determined in this way was normalized with the maximum reaction rate.

It can be seen from FIG. 1 that the reaction rate of the citral hydrogenation low citral concentrations increases hyperbolically with the citral concentration, but then decreases sharply at higher citral concentrations. The maximum V _{max of} the reaction rate is reached at a citral concentration of approximately 9% by weight. Reaction rates that are equal to 0.8 times V _max or higher are achieved in the concentration range from about 5 to 20% by weight of citral.

2, a plant suitable for carrying out the method according to the invention comprises a first hydrogenation reactor 201 and a second hydrogenation reactor 202.

The hydrogenation reactor 201 comprises a stirrer 203. A liquid comprising a starting material to be hydrogenated is fed to the hydrogenation reactor 201 via line 204. Hydrogen gas is fed to the hydrogenation reactor 201 via line 205. The exhaust gas from the hydrogenation reactor 201 is discharged from the hydrogenation reactor 1 via line 206. The discharge stream 207 carried out at the bottom of the hydrogenation reactor 201 is fed to the hydrogenation reactor 202. The hydrogenation reactor 202 comprises a stirrer 208. Hydrogen gas is fed to the hydrogenation reactor 202 via line 209. The exhaust gas from the hydrogenation reactor 202 is discharged from the hydrogenation reactor 202 via line 210.

The hydrogenation reactor 202 comprises built-in trays 21 1, which limit backmixing in the hydrogenation reactor 202. The floors are preferably perforated sheets. Dispersed hydrogen gas is at least partially retained by the trays, so that the proportion of dispersed gas in the liquid reaction mixture drops in the section located at the outlet from the second reactor. At least at the outlet, the hydrogenation takes place in a liquid single phase.

A liquid containing the hydrogenated reaction product is discharged from the hydrogenation reactor 202 via line 212.

A liquid comprising a homogeneous rhodium catalyst which has at least one chiral ligand is metered into the liquid stream 204 (not shown).

According to FIG. 3, a plant suitable for carrying out the method according to the invention comprises a first hydrogenation reactor 301 and a second hydrogenation reactor 302. The hydrogenation reactors 301 and 302 are mixed with injection and each have a pumping circuit.

The hydrogenation reactor 301 comprises a jet nozzle 303. A liquid from line 304 is fed to the hydrogenation reactor 301 via the jet nozzle 303. The liquid stream from line 304 comprises the educt stream 305 and the pumping stream 306. The educt stream 305 comprises a liquid comprising an educt to be hydrogenated. The pumping stream 306 is a partial stream of the discharge stream 307 carried out at the bottom of the hydrogenation reactor 1. Hydrogen gas is fed to the hydrogenation reactor 301 via line 308. The hydrogenation reactor 301 comprises a guide tube 309, the jet nozzle 303 being arranged above the liquid level and the gas / liquid jet generated by the jet nozzle 303 being directed into the guide tube 309. The exhaust gas from the hydrogenation reactor 301 is discharged from the hydrogenation reactor 301 via line 310.

A partial stream of the discharge stream 307 is fed in with the liquid pumping stream 311 withdrawn from the backmixed zone of the hydrogenation reactor 302 Liquid stream 312 is combined and fed to the hydrogenation reactor 302 via the jet nozzle 313. Hydrogen gas is fed to hydrogenation reactor 302 via line 314. The hydrogenation reactor 302 comprises a guide tube 315, the jet nozzle 313 being arranged above the liquid level and the gas / liquid jet generated by the jet nozzle 313 being directed into the guide tube 315. The exhaust gas from the hydrogenation reactor 302 is discharged from the hydrogenation reactor 302 via line 316.

The hydrogenation reactor 302 comprises built-in trays 317 which limit backmixing in the hydrogenation reactor 302. The floors are preferably perforated sheets. Dispersed hydrogen gas is at least partially retained by the trays, so that the proportion of dispersed gas in the liquid reaction mixture drops in the section located at the outlet from the second reactor. At least at the outlet, the hydrogenation takes place in a liquid single phase.

A liquid containing the hydrogenated reaction product is discharged from the hydrogenation reactor 302 via line 318.

A liquid comprising a homogeneous rhodium catalyst which has at least one chiral ligand is metered into the pumping circuit of the hydrogenation reactor 301, for example into the liquid stream 304 (not shown). The pumping circuit of the hydrogenation reactor 301 and the hydrogenation reactor 302 comprises an external heat exchanger (not shown), which cools the discharge stream 307 and the pumping stream 311, for example.

example

1,500 kg / h of citral and 1,500 kg / h of a catalyst mixture were fed into a jet loop reactor with a liquid volume of 12 m ³ . The catalyst mixture was a mixture produced analogously to the instructions described in US 2018/057437 A1, WO 2006/040096 and WO 2008/132057 A1, Rh (CO) 2acac, Chiraphos and Tridodecylamine in a molar ratio of 1: 1, 4 : 10 were implemented in Citronellal with CO and H ₂ . The concentration of the catalyst mixture in the jet loop reactor was 300 to 1,000 ppm by weight, based on the amount of rhodium in the catalyst. The ratio of external circulation to inflow was 380: 130. The ratio of internal circulation to inflow was 3,000: 1000. The power input was 2 kW / m ³ . The pressure in the reactor was regulated to 80 bar by feeding in hydrogen gas (containing 1,000 ppm carbon monoxide). The temperature in the reactor was regulated at 22 ° C. Citral sales were 88%. A citral concentration of about 7% by weight was present in the discharge from the first reactor.

The reactor discharge was fed into a second reactor with a liquid volume of 9.7 m ³ . The pressure in the second reactor was regulated to 80 bar by feeding in hydrogen gas (containing 1,000 ppm carbon monoxide). The temperature in the reactor was regulated at 22 ° C. The total conversion after the second reactor was between 93 and 99.9%.

It was shown that the reaction rate in the first reactor decreased at higher citral concentrations. It thus proved advantageous to operate the first reactor at low citral concentrations and to complete the reaction in the second reactor.

Claims

Claims 1. Process for the continuous production of an optically active  Carbonyl compound by asymmetric hydrogenation of a prochiral a, b-unsaturated carbonyl compound with hydrogen in the presence of a homogeneous rhodium catalyst which has at least one chiral ligand, wherein a liquid reaction mixture containing the prochiral a, b-unsaturated carbonyl compound in a first, back-mixed reactor undergoes gas / liquid two-phase hydrogenation, and the liquid reaction mixture is then further hydrogenated in a second reactor, with a concentration of the prochiral in the first reactor  a, b-unsaturated carbonyl compound from 3 to 20 wt .-%. 2. The method of claim 1, wherein one to enter the second  The reactor section of the second reactor disperses hydrogen gas in the liquid reaction mixture. 3. The method according to claim 1 or 2, wherein one in the first reactor Concentration of the prochiral a, b-unsaturated carbonyl compound at which the reaction rate is at least 0.8 times V _max and V _{max is} the maximum reaction rate when the  Reaction rate against the concentration of prochiral  a, b-Unsaturated carbonyl compound. 4. The method according to any one of the preceding claims, wherein in the  second reactor the liquid reaction mixture up to a concentration of the prochiral a, b-unsaturated carbonyl compound of less than 5 wt .-%. 5. The method according to any one of the preceding claims, wherein the ratio of the reaction volume of the first reactor to the reaction volume of the second reactor is 1: 1 to 1: 5. 6. The method according to any one of the preceding claims, wherein the first reactor is characterized by a number of vessels N in the range from 1 to 3. 7. The method according to any one of the preceding claims, wherein the volume-specific power input in the first reactor is 0.5 to 5 kW / m ³ . 8. The method according to any one of the preceding claims, wherein the first reactor is designed as a loop reactor. 9. The method according to any one of the preceding claims, wherein one  Backmixing in the second reactor is limited and the hydrogenation is carried out in a liquid-single phase at least in a section of the second reactor which is located at the outlet from the second reactor. 10. The method of claim 9, wherein the backmixing in the second reactor is limited by internals. 1 1. The method according to any one of the preceding claims, wherein the second reactor is characterized by a boiler number N of more than 4. 12. The method according to any one of the preceding claims, wherein the prochiral  a, b-unsaturated carbonyl compound is selected from compounds of the general formula (I)

 wherein R ¹ , R ² are different from one another and each represent an unbranched, branched or cyclic hydrocarbon radical having 1 to 25 carbon atoms, which is saturated or has one or more, non-conjugated ethylenic double bonds, and which is unsubstituted or one or more of the same or carries various substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms, R ^{3 represents} hydrogen or an unbranched, branched or cyclic hydrocarbon radical having 1 to 25 carbon atoms, which is saturated or one or more, non-conjugated has ethylenic double bonds, and which is unsubstituted or carries one or more identical or different substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms, or R ³ together with one of the radicals R ¹ or R ² also a 3 to 25-membered alkylene group in which 1, 2, 3 or 4 non-adjacent Chh groups can be replaced by O or NR ^5c , where the alkylene group is saturated or has one or more non-conjugated ethylenic double bonds, and the alkylene group is unsubstituted or carries one or more identical or different substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, Ci-C4-alkyl, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms, with two substituents also together for can be a 2 to 10-membered alkylene group, the 2 to 10-membered alkylene group being saturated or having one or more non-conjugated ethylenic double bonds, and the 2 to 10-membered alkylene group being unsubstituted or one or more of the same or carries various substituents which are selected from OR ⁴ , NR ^5a R ^5b , halogen, C ₆ -Cio-aryl and hetaryl with 5 to 10 ring atoms; in which R ^{4 represents} hydrogen, Ci-C ₆ alkyl, C ₆ -Cio-aryl, C6-Ci4-aryl-Ci-Cio-alkyl, or Ci-Cio-alkyl-C6-Ci4-aryl-; R5 ³ , R5b each independently of one another are hydrogen, Ci-C ₆ -alkyl, OQ-O-IO-aryl, C ₆ -C 14 -aryl-C i-Cio-alkyl or C ₁ -C ₁₀ -alkyl-Ce-C ₁₄ -aryl mean or R ^5a and R ^5b together can also mean an alkylene chain with 2 to 5 carbon atoms, which can be interrupted by N or O; and R ^{5c represents} hydrogen, Ci-C ₆ alkyl, C ₆ -Cio-aryl, C6-Ci4-aryl-Ci-Cio-alkyl or Ci-Cio-alkyl-C6-Ci4-ary-. 13. The method according to claim 12 for the production of optically active citronellal of the formula (III)

where ^* denotes the center of asymmetry; by asymmetric hydrogenation of geranial of the formula (la-1) or of neral of the formula (lb-1)

 (la-1) (lb-1) or a mixture containing neral and geranial. 14. The method according to any one of the preceding claims, wherein the  Catalyst concentration is 0.001 to 1 mol% based on the amount of prochiral a, b-unsaturated carbonyl compound in the reaction mixture, calculated as rhodium atoms contained in the catalyst. 15. The method according to any one of the preceding claims, wherein the chiral ligand is a chiral bidentate bisphosphine ligand, in particular Chiraphos. 16. The method according to any one of the preceding claims, wherein the method is carried out in the presence of a compound of the formula (II),

in which Z in formula (II) represents a group CHR ³ R ⁴ and in which the variables R ¹ , R ² , R ³ , R ⁴ independently of one another and in particular together have the following meanings: R ¹ , R ² : are the same or different and represent phenyl, that is unsubstituted or carries 1, 2 or 3 substituents which are selected from methyl and methoxy, where R ¹ and R ^{2 in} each case are in particular unsubstituted phenyl; R ³ represents C to C ₄ alkyl, in particular methyl; R ⁴ stands for C to C ₄ alkyl which carries a group P (= 0) R ^4a R ^4b and in particular a group CH ₂ -P (= 0) R ^4a R ^4b or CH (CH ₃ ) -P (= 0) R ^4a R ^4b ; in which R ^4a , R ^4b : are the same or different and represent phenyl, that  is unsubstituted or carries 1, 2 or 3 substituents selected from Methyl and methoxy, with particular preference R ^4a and R ^4b each representing unsubstituted phenyl. 17. A process for the production of optically active menthol, in which optically active citronellal of the formula (III) is prepared by a process of claims 13 to 16, the optically active citronellal of the formula (III) undergoes cyclization to optically active isopulegol and the optically active isopulegol hydrogenated to optically active menthol.